safe boat repair

[_Dom]

I have a project which involves welding on a 28' aluminum boat with integral fuel tank. In my business we take Safety very seriously. This welding repair is within 3' of the fuel filler/vent, tank capacity 150 gal. Planning this work includes written documentation and daily safety briefing before any welding begins. I have pumped out usable fuel through the main line by disconnecting at filter inlet. No doubt some fuel remains in the tank. I plan on purging the (mostly) empty tank with nitrogen and monitoring the space close to welding operations with a four gas monitor.

This picture shows my nitrogen connection to fuel tank inlet and PVC pipe taped around the vent which directs the flow up eight feet. In addition I have an oxygen sensor which I can use in the vent stack at the bottom to verify no oxygen present. I was asked by our Safety if pockets of explosive vapors remain in the tank due to internal baffles or other unknown design features. What do you think?

[_Dom]

This monitor setup shows the stack contains enough oxygen to be an explosive hazard with nitrogen flowing. By opening the threaded ABS collar I can sample oxygen at the base of the stack, it was less than 8%. I interpret that to mean the tank space involving filler and vent is not a hazard, no oxygen present. My question involves the design of a typical marine fuel tank. Could this tank have trapped pockets of explosive vapor due to baffles?

[Capt B]

I would say so most marine tanks I have dealt with are baffled. It is unlikely you got all the fuel out. I do a good bit of welding on aluminum boats and usually remove the fuel tank. Obviously this isn't an option for you. The couple times I have had to leave a fuel tank in the boat. I have plugged all fuel fittings and ran a long vent clear of the area. Probably not the smartest way to do it. You're plan of purging sounds better but I am no expert on fuel tanks.

[rlitman]

Can you get a camera in there? Yes, most marine tanks are baffled. Both to impede sloshing, and for structural strength. But I can't imagine that the upper edge of the baffles are completely sealed. That would trap air in a tank, preventing it from filling. Baffles usually have holes, and that should let the air out.

[Yofish]

What EXACTLY are you welding? If it's a void, you better void the void!!!It looks to me that the tank is not integral to the hull but to the deck? I know of four skiffs that were completely destroyed from welding on a void. A void ANYWHERE in a skiff MUST be treated like it is a fuel tank on as gasoline powered watercraft; I don't care how far away from the fuel source. Worry about that more than the tank itself!

[ezduzit]

The baffles will not prevent complete emptying of tank.

[Yofish]

About emptying a fuel tank - to begin with - THAT is an impossibility without aeration! There will always be residual that can only be remover by aeration, whether over time or by force, baffles only make that a bit more difficult. I've posted this too many times but I'll do it again because this is something that must be approached cautiously. I have a small squirrel cage blower with a flex hose that I pump warm air into the tank after removing all the fittings to provide escape. The fact is, a full tank is safer than one with a cup for gas in it.

This is a true story: I had a weldor bud that built a seine skiff with a combo tank/seat on the deck and the client wanted a mod to the tank. OK, sez weldor bud, bring it in with the tank full. He was just about done welding an angle to the tank (1/8" 5052) and he punched through a smidge. Burned the whole thing up and had to build a whole knew skiff! No explosion.

Last year, some youngsters (that I've been teaching skiff building to) were doing a repair on a set-net skiff transom; the transom was about six feet from a (integral to the hull) 40 gal gas tank that was full. The instant they struck an arc the whole thing blew, completely pumkinizing the thing and splitting it in half, dumping all the fuel on the floor but it didn't ignite. I'm not sure they actually know how lucky they were, especially in that their shop is spray-foamed without covering.

A little highly reactive fuel and lots of available oxidizer in a confined space is a bomb, it's that simple.

[Kevin Morin]

_Dom,
the marine tank needs to vent fully to fill. If the filling liquid does not have a series of openings in the baffles that would vent all gas, then gas trapped over the liquid would remain: and the tank would not fill due to the trapped gases you're inquiring about.

Therefore-any marine tank that does fill: Cannot have any trapped vapors by the baffles- otherwise it would not fill as those vapors/gases/atmosphere would not vent allowing liquid to fill the volume of these-trapped gases.

So, there is no way to trap vapors with correctly built baffles as they are all 'limbered'/moused holed/clipped at the sides/vented to allow ALL gas to escape during filling. If you have not confidence the tank is well built- this information is of little or no use in your considerations.

I am not discussing tank purging or inerting, or safety precautions needed to work on this skiff- I'm addressing the logic of your question about trapped vapors; ONLY.

cheers,
Kevin Morin
Kenai, AK

[_Dom]

CaptB - Agreed it is unlikely I got all the fuel out. Am considering plugging all lines like you mentioned, thinking twin 115 Hp outboards will be lots of work to remove but if it keeps me from frying any electronics it will be worth it. Long vent clear of the area is my 8' vertical stack but my connections are not gastight, may be wise to connect directly to the vent connection at top of the tank.

rlitman - I thought of a camera, went looking for a Milwaukee M-Spector i have around here somewhere...
https://www.redtoolstore.com/product...xoCiBcQAvD_BwE
Problem with this (besides not having seen that camera in a few years) is I tried to fish a siphon hose into the tank through the filler with no luck. Might disconnect the fill hose at tanktop and see what it looks like.

Yofish - This repair is limited to gunwale, 6-8" above deck. The center console is three feet away, filler and vent are there. fuel tank in the void below deck, a few inches clear space between. Will sketch something a bit more descriptive and attach. I will start thinking about the void between deck and hull as a fuel tank to be monitored. Great story about the transom 6' away and resulting explosion, exactly the story I need to consider.

Kevin - Thanks for addressing the baffles question, I do have some confidence in the construction. Regarding the safety considerations, i am trying to understand all aspects before hiring a local Marine Chemist who will come out and certify the repair is ready to weld. My theory when hiring professional services is to know in advance what is expected. I have collected many opinions on this repair, all contributions greatly appreciated.

[_Dom]

ezduzit - I understand, more curious if a marine tank might typically have the outlet at lowest point. Maybe I need to adjust the trim and heel as it sits in the shop to attempt draining the last drop of fuel. Most guidance I receive indicates I should consider it more dangerous than a full tank which makes sense to me.

[ezduzit]

... curious if a marine tank might typically have the outlet at lowest point. Maybe I need to adjust the trim and heel as it sits in the shop to attempt draining the last drop of fuel...

Gas tanks will have their outlet virtually directly above the lowest point, leading out the top of the tank. Whereas a diesel tank can have the outlet directly in the bottom of the tank and at the lowest point.

The baffles will not prevent fuel from running to the lowest point. But it is obviously important for you to trim the level of the tank so you can get the fuel out. I have heard of gas tanks being filled with water, to displace latent fumes, for tank welding repairs. However it seems unnecessary, in this case, if you are not welding on the actual tank.

[_Dom]

Seems obvious now that we are talking about it but I hadn't considered it until you mentioned it. Will check trim and heel on the tank top tomorrow, adjusting as required. No quarantee the top is parallel the bottom but at least I can access it.

[Yofish]

A teaspoon of gasoline vs. a teaspoon of gunpowder: Gasoline has a much higher energy density than gun powder BUT does not carry its own oxidizer thus the reason a little gas and enough oxygen is dangerous when confined and the reason I stated that a full tank is less likely to blow than an empty one. Also the reason that it's easy to see a teaspoon of petrol in a void can be rather dangerous. A gram of petrol needs about 15 grams of air for complete combustion. Note that INCOMPLETE combustion can be bad enough!

I'm not you and I'm not where you are staring at this problem but I can tell you that I've welded many times on skiffs that far away from a void and never blown myself up. I use a blanket to catch any berries that might land inside and just go for it but, that's me. If the repair was crossing a void - !never! - without furthertime consuming precautions. A tank 3' away from the repair zone wouldn't bother me and I'd prefer it be full, then top 'er off with C02 or whatever inert gas is on hand and plug the vent afterwards.

Show us a pic of EXACTLY what needs repair, por favor.

[frieed]

...I plan on purging the (mostly) empty tank with nitrogen and monitoring the space close to welding operations with a four gas monitor...

any chance you meant Argon instead of Nitrogen? Nitrogen is lighter than air and won't purge the bottom of the tank.

Molecular weights:
Nitrogen 28
Air 29
Argon 39

[_Dom]

Yofish - I hear you about wishing the tank was full, pumped out 50 gallons and considered re-filling it. Once I realized it had 150 gal capacity it seemed better to leave it as close to empty as possible. All available experts agree distance from the void space is adequate, I dreamed up this purge idea considering the risk of a "mostly" empty tank. Pictures show the center console in relation to side of the boat with holes to be patched and looking through the deck hatch at bottom and top of fuel tank so you can better see the repair.

frieed - I have nitrogen readily available. Could use Argon, we have a few cylinders available. Thanks for reminding me N2 is lighter. Maybe I can plumb the nitrogen into the fuel tank outlet instead of the filler. I understand this is not a complete solution, would start the purge with Argon and switch over. I have plenty of N2, argon supply is more limited.

[Kevin Morin]

_Dom,

the corrosion on this boat is so totally widespread its hard to understand!!! ?? was it stored in a container? where there was humidity but limited drying? Those white flowers are mill scale crevice corrosion sites and they're present in every shot you've posted!

This boat needs an acid wash to etch the surface . The oval repair pieces confirm my earlier prediction that a strip above the deck is the correct way to do the repair and will in the end work out to be the least cost and effort- but your short sighted 'owner' will probably not see that until the four layers of work, redundant coatings, cleanings, and all that remains to mount a bumper is spent!

Always funny to those of us who build- kind of gallows humor at the expense of the inexperienced- but still amusing to see someone "saving money" by only spending 20% payments six or eight times- instead of doing the job right - once!

But then, living and learning and all that. Tank wall corrosion is starting, like the under deck (not likely vented properly?) areas and structurals all showing corrosion means that no one who understood aluminum boats has EveR owned this boat.


If you have a bore scope- it will be worth the time to inspect the tank (i'm not addressing safing that void to weld) to see the extent of the water bottom damage already in progress- my ***-umption is based on the super deteriorated condition of the rest of the boat. Anyone who'd allow the boat to get that 'far gone' won't even think of tank internals' corrosion from water bottoms going acidic but remaining inside an unetched tank!

Don't remotely envy you having to work on this boat, as Yofish has remarked before (not necessarily on this thread) I'd have turned down the work - no sense in adding my name to the roster on this project. Too many mistakes in the past to become the "last man" in line of errors!!

Cheers,
Kevin Morin
Kenai, AK

[_Dom]

Kevin - Long winded response moved to other thread in attempt at keeping this focused on Safety. Appreciate your comments, as I run out of excuses to strike an arc I seem to get more concerned about potential fallout.

[Yofish]

For us that make these sorts of things, it would be cool to see some more pics (even though it be off-topic) of the overall hull. Can't figure what the hell the thing is for. Those low sides (?) and strange extreme outboard bollards have me puzzled. What need is there to cover those cut-outs?

As Kevin points out, the hull is a mess. It looks like the Freeman hatch cover wasn't even used or didn't have a gasket by the looks of below decks. Any idea when it was built?

[_Dom]

Basically a square nosed, flat bottom workboat 28'. Used for working around the pier repositioning oil containment boom, etc. When the marine crew doesn't have access to this (before 2004 and now over a year) they drop their fast rescue boat from the ship. Can't see it in this picture but that bollard was completely broken off sometime ago. Repair weld belongs in the welding fail section, if I attempt to rework that mess I might take some pictures. A few other areas need attention as you can imagine. My focus now is completing the job without joining the Darwin hall of shame. If I totally goober the welding I am getting lots of practice with a grinder.

[Kevin Morin]

_Dom
If I may offer some comments on the gas monitoring method you show using the 'Clip' type of hook up and 'face fitting' with tube? My comment is this doesn't allow a real life full flow test of the actual atmosphere in the tank.

To make a real flow reading I'd want a T to atmosphere before the test instrument. BW clips are intended to be 'walked around' and open to the air- to test a continuously shifting concentration- this method of sampling (shown) is just slightly suspect- not only because the tube is almost "dead legged" to the instrument- thereby reducing real time flowing concentrations but.... the instrument isn't really working in its own design environment. The hand bulb only really pulls new gases out on demand and that isn't a really accurate way to monitor concentrations in the tank?

By using the T to purge the line- and flush old gas out and new in- the set up would be more likely to give a true value of gas concentrations in the sampled volume.

However, the full sized (BW or other product [MSA etc.]) gas monitoring unit has a built-in battery driven pump- just exactly for this purpose- to constantly move gases at decent velocities through the monitoring cells' cavities in the instrument's head/cap in order to allow the changes in gases' concentrations to shift, in real time, as they do in the purged volume.

Also not sure if you're monitoring LEL (lower explosive limits) or O2 just the presence of oxygen in the tank being sampled? The LEL seems the safer test of air/gases in the tank compared to just testing O2?

So, while I'm splitting hairs to some degree- I have been BW's Alaskan Sales rep in my past, and did run and instrument and controls company for a couple decades so I've run into various sampling/testing set ups and wanted to share the questions based on your photo's posted.

I think you've mentioned you were arranging an external tech to come monitor or help monitor the tank so my remarks may not be germane? Just another note about "If I were in your shoes"...

Cheers,
Kevin Morin
Kenai, AK

[_Dom]

Thanks for all your comments Kevin, greatly appreciated. The photo was taken early in my learning curve trying to understand best practices. I realize the sampling method is suspect and continue to refine my approach. Looking online I think MSA Altair 5x or BW GasAlert XT are the battery powered instruments you mention, my best approach for something in that price range will be to hire a local expert, relying on his equipment and expertise in this area.
I originally thought to use a four gas monitor in the space between hull and deck and the oxygen monitor to verify no oxygen at my cobbled up sample port in the vent stack. I have been re-thinking my setup and at this time trying to weigh the advantages/disadvantages of running a continuous nitrogen purge compared to filling it with argon and sealing it up.
Hand bulb as shown was available to me, so I can use it to test various arrangements before employing the battery operated instrument. I wonder if the "T" you mention would be appropriate with both the battery pump and the hand bulb? Not at all clear how that would improve my process, is this sketch what you have in mind?

[_Dom]

Reviewing the conversation so far it occurs to me I may have used the wrong terminology. This fuel tank is located between hull bottom plates and deck. Possibly not "integral" to either, it is three feet away from the sides where welding will take place. Not part of the discussion about purging the fuel tank, just wanted to clarify.

[prconnection]

Wow.....just read this whole thread. Is the client being billed hourly for this job? Is the only welding taking place on the sides of the boat that u showed in the picture (putting fillers in the side of the boat above the deck)?

Just curious becuase if that is the only welding taking place, nothing on the deck, nothing under the deck and nothing on the fuel tank I am puzzled to all the fuss.The fuel tank looks like a standard welded aluminum tank that is hung under the floor like 90 percent of the welded aluminum boats. If the tank isn't leaking into the hull I don't see anything extra ordinary to be dealing with. Grab a flap disk clean all the corrosion where the weld is to take place and weld the patches in place. Put ground clamp as close to the weld site as possible and done.

Seems like a lot of whoopla for something so simple unless I'm not understanding exactly what is being welded

Our shop here welds on many many many aluminum boats. We do not weld on fuel tanks of any kind that have had gas in them,unless they are brand new, never been filled. If they are brand new it's becuase we made it. We don't weld on a hull if the tank is leaking.......pretty easy to tell if it leaks as it reeks of gasoline. Other then that we just get at it.

[Kevin Morin]

_Dom, I do understand and the tank's images were clear the location and relative proximity to the weld areas.

prcon..... , since your firm guys doesn't weld on (near, around, hulls with...? ) tanks that are used.... maybe you might be more concerned if one propagated some O2 reduction? The issue is High Frequency associated with TIG. MIG doesn't use an arc stabilizer that wanders all over the place... so it won't be a risk if the tank vents and fittings are all covered- however with TIG.... an LEL mixture inside the tank could be a problem.

Sometimes the most well behaved of all welding arc stabilizers act un-explained: but then... some folks have had what we might call volatile experiences - maybe we could call them volcanic experiences with TIG high freq around vapor filled tanks.

this includes welders who were welding 20 feet from a tank on the hull and the tank went up- I don't understand high frequency well enough to understand how or why is acts as it does but making sure the vapors in a tank , welded into a hull that will be welded on- is just plain good practice.

So far in the last 40some years on welded aluminum boats; I've: fried other welders' controls with TIG hi-freq, ruined an expensive radar, two inboard engine 'brains' and one outboard control module; lit up two tanks- they just popped and were both in the same boat (both bolted into the bilge) and managed to ruin the radio reception of the whole shop for a long time.

Boy! that has been expensive and when those tanks popped- it scared the "stuffing" out of me as I was under a lazarette deck working on the rudder mounts- 6' from the emptied tanks (purged I thought!!! ?) and the ground/work clamp was under the weld- about 3" below the fitting I was TIG welding.

In my experience, being careful of TIG hi-freq is just good practice.

So back to 'safing' the tank - the sample system you show, _Dom, allows dead legs so the actual current vapor composition is not 'true' in all cases. The electric pump type gas composition monitors vents the sample continuously so - if it reads "All Clear" on an LEL sensor- you can weld without risk. If the sampling of a void is not done by full time flowing electric pump sampled flow- and that flow vented off the sensors' surfaces- then the actual reading is not reflective of the composition inside the void being sampled.

I don't think there's a high risk of igniting the tank as shown, but, making sure there is no LEL showing in a valid sampling system- is just plain common sense, IMO?

Cheers,
Kevin Morin
Kenai, AK

[prconnection]

Thanks Kevin....... usually but not always repairs on these boats are done with a miller pulse mig. It's rare we tig on the boat itself. In general all new parts are tiggged on the bench and migged in place. Of course there are exceptions.

I do very much thank you for the high frequency tig information. Definitely something to thing of.